Benchmarks / Artificial Analysis

Reported by Artificial Analysis

Artificial Analysis

Share of questions answered correctly over long documents, as run by Artificial Analysis with its own harness and prompts.

Last updated 20 Oct 2026

Results dated
20 Oct 2026
Results
399 configurations of 239 models
Unit
% of questions
Licence
Artificial Analysis commercial data licence

Long-context reasoning (AA-LCR): Nemotron 3.5 Super VL

Top 15 of 239 results · % of questions, higher is better. Choose a model to highlight it.Clear highlight

  1. 1 Kimi K3 (max reasoning)Moonshot AI 88.7%
  2. 2 MiMo-V2.6-ProXiaomi 86.3%
  3. 3 Claude Fable 5.1 (max reasoning)Anthropic 85.3%
  4. 4 Claude Opus 5.5 (max reasoning)Anthropic 84.7%
  5. 5 GPT-5.5 (high reasoning)OpenAI 84.3%
  6. 6 DeepSeek-V4.1-Flash (max reasoning)DeepSeek 84.0%
  7. 6 Gemini 3.8 Flash (medium reasoning)Google 84.0%
  8. 6 GPT-5.6 Sol (max reasoning)OpenAI 84.0%
  9. 6 GPT-6.1 Sol (low reasoning)OpenAI 84.0%
  10. 10 GPT-5.6 Luna (max reasoning)OpenAI 83.7%
  11. 10 GPT-6 Sol (high reasoning)OpenAI 83.7%
  12. 12 Muse Glimmer (high reasoning)Meta 83.3%
  13. 12 GPT-5.3-Codex (extra-high reasoning)OpenAI 83.3%
  14. 12 GPT-6 Luna (max reasoning)OpenAI 83.3%
  15. 15 Gemini 3.7 Flash (medium reasoning)Google 83.0%
  16. 121 Nemotron 3.5 Super VLNVIDIA 69.0%

Full results

Artificial Analysis: Long-context reasoning (AA-LCR), % of questions, higher is better
#ModelLong-context reasoning (AA-LCR)
% of questions, higher is better		Price
$ per million tokens, in / out
1 Kimi K3 (max reasoning)Moonshot AI · best of 2 settings
88.7%
$3 / $15
2 MiMo-V2.6-ProXiaomi
86.3%
$0.43 / $0.87
3 Claude Fable 5.1 (max reasoning)Anthropic · best of 5 settings
85.3%
$10 / $50
4 Claude Opus 5.5 (max reasoning)Anthropic · best of 5 settings
84.7%
$4 / $20
5 GPT-5.5 (high reasoning)OpenAI · best of 5 settings
84.3%
$5 / $30
6 DeepSeek-V4.1-Flash (max reasoning)DeepSeek · best of 2 settings
84.0%
$0.30 / $1.20
6 Gemini 3.8 Flash (medium reasoning)Google · best of 3 settings
84.0%
$1.50 / $7.50
6 GPT-5.6 Sol (max reasoning)OpenAI · best of 6 settings
84.0%
$4 / $20
6 GPT-6.1 Sol (low reasoning)OpenAI · best of 5 settings
84.0%
$2 / $10
10 GPT-5.6 Luna (max reasoning)OpenAI · best of 6 settings
83.7%
$0.20 / $1.20
10 GPT-6 Sol (high reasoning)OpenAI · best of 6 settings
83.7%
$2 / $10
12 Muse Glimmer (high reasoning)Meta
83.3%
–
12 GPT-5.3-Codex (extra-high reasoning)OpenAI
83.3%
$1.75 / $14
12 GPT-6 Luna (max reasoning)OpenAI · best of 6 settings
83.3%
$0.10 / $0.50
15 Gemini 3.7 Flash (medium reasoning)Google · best of 3 settings
83.0%
$1.50 / $7.50
15 Muse Spark 1.3 (max reasoning)Meta · best of 2 settings
83.0%
$1.25 / $4.25
15 MiniMax-M3MiniMax
83.0%
$0.30 / $1.20
15 GPT-5.6 Terra (max reasoning)OpenAI · best of 6 settings
83.0%
$2 / $12
19 Claude Haiku 5.5 (max reasoning)Anthropic · best of 5 settings
82.7%
$0.10 / $0.50
19 Claude Sonnet 5.5 (max reasoning)Anthropic · best of 5 settings
82.7%
$2 / $10
19 DeepSeek-V4.1-Flash-1017 (max reasoning)DeepSeek
82.7%
$0.30 / $1.20
19 GPT-5.2 (extra-high reasoning)OpenAI · best of 3 settings
82.7%
$1.75 / $14
23 Claude Fable 5 (max reasoning)Anthropic
82.3%
$10 / $50
23 GPT-5.2-Codex (extra-high reasoning)OpenAI
82.3%
$1.75 / $14
25 Qwen3.8-27B (extra-high reasoning)Alibaba · best of 4 settings
82.0%
$0.50 / $3
25 Claude Opus 5 (medium reasoning)Anthropic · best of 5 settings
82.0%
$5 / $25
25 Claude Sonnet 5 (max reasoning)Anthropic · best of 6 settings
82.0%
$2 / $10
25 Gemini 3.1 Pro PreviewGoogle
82.0%
$2 / $12
25 GPT-5.4 (extra-high reasoning)OpenAI · best of 3 settings
82.0%
$2.50 / $15
30 DeepSeek-V4-Flash-Vision-Exp (max reasoning)DeepSeek
81.3%
$0.44 / $1.32
30 Mistral Large 4Mistral AI
81.3%
$1.36 / $4.18
32 Kimi K2.6Moonshot AI · best of 2 settings
81.0%
$0.95 / $4
32 Grok 4.6 (medium reasoning)xAI · best of 4 settings
81.0%
$2 / $6
34 Qwen3.6-Max-PreviewAlibaba
80.7%
$1.03 / $6.16
34 Muse Wonder (max reasoning)Meta
80.7%
$4 / $20
34 GPT-6 Astra (max reasoning)OpenAI · best of 5 settings
80.7%
$10 / $50
37 Qwen3.8-2.4T-A95BAlibaba
80.3%
$2 / $6
37 Qwen3.8-Max (0902)Alibaba
80.3%
$2 / $6
37 DeepSeek-V4-Pro (0813, max reasoning)DeepSeek · best of 2 settings
80.3%
$1.32 / $3.96
40 Claude Sonnet 4.6 (max reasoning)Anthropic · best of 2 settings
80.0%
$3 / $15
40 Gemini 3.6 Flash (high reasoning)Google
80.0%
$1.50 / $7.50
40 GPT-5.1 (high reasoning)OpenAI · best of 2 settings
80.0%
$1.25 / $10
40 Grok 4.7 (low reasoning)xAI · best of 3 settings
80.0%
$2 / $6
40 GLM-5.3-FlashZ.ai
80.0%
$0.15 / $0.50
45 Qwen3.8-Flash-NextAlibaba
79.7%
–
45 DeepSeek-V4-Flash (0731, max reasoning)DeepSeek
79.7%
$0.14 / $0.28
45 Gemini 4 Argon (high reasoning)Google
79.7%
$2 / $10
45 MiniMax M3.1 FlashMiniMax
79.7%
$0.15 / $0.60
45 MiMo-V2.5-Pro (reasoning on)Xiaomi · best of 2 settings
79.7%
$0.43 / $0.87
45 GLM-5.3 (max reasoning)Z.ai · best of 2 settings
79.7%
$1.40 / $4.40
51 Kimi K2.7 CodeMoonshot AI
79.3%
$0.95 / $4
51 Grok 4.5 (high reasoning)xAI
79.3%
$2 / $6
53 Qwen3.7-MaxAlibaba
79.0%
$1.48 / $4.42
53 Muse Spark 1.2 (extra-high reasoning)Meta
79.0%
$1.25 / $4.25
53 Hy3Tencent
79.0%
$0.14 / $0.58
56 Claude Opus 4.7 (max reasoning)Anthropic · best of 2 settings
78.7%
$5 / $25
57 Qwen3.6-PlusAlibaba
78.3%
$0.33 / $1.95
57 Qwen3.8-Max (0803)Alibaba
78.3%
–
57 Muse Spark Open (extra-high reasoning)Meta
78.3%
–
57 MiniMax-M2.7MiniMax
78.3%
$0.30 / $1.20
57 GLM-5.2 (max reasoning)Z.ai · best of 2 settings
78.3%
$1.40 / $4.40
62 GPT-5 (high reasoning)OpenAI · best of 2 settings
78.2%
$1.25 / $10
63 Claude Opus 4.6 (max reasoning)Anthropic · best of 2 settings
78.0%
$5 / $25
63 Gemini 3 Flash Preview (reasoning on)Google · best of 2 settings
78.0%
$0.50 / $3
63 Muse SparkMeta
78.0%
–
63 Kimi K2.5Moonshot AI · best of 2 settings
78.0%
$0.57 / $2.85
63 Beam 1.0 501B A23BReflection
78.0%
–
68 Qwen3.5-27BAlibaba · best of 2 settings
77.7%
$0.27 / $2.16
68 Claude Opus 4.8 (max reasoning)Anthropic
77.7%
$5 / $25
68 Muse Spark 1.1 (extra-high reasoning)Meta
77.7%
$1.25 / $4.25
71 Qwen3.5-397B-A17BAlibaba · best of 2 settings
77.3%
$0.55 / $3.50
71 Qwen3.6-27BAlibaba · best of 2 settings
77.3%
$0.30 / $3.20
71 Claude Opus 4.5 (reasoning on)Anthropic · best of 2 settings
77.3%
$5 / $25
71 Inkling (extra-high reasoning)Thinking Machines
77.3%
$0.95 / $4.05
75 GPT-5.4 mini (extra-high reasoning)OpenAI · best of 3 settings
77.0%
$0.75 / $4.50
76 GPT-5.4 nano (extra-high reasoning)OpenAI · best of 3 settings
76.7%
$0.20 / $1.25
77 Qwen3.5-122B-A10BAlibaba · best of 2 settings
76.3%
$0.26 / $2.08
78 Claude Opus 4.1 (reasoning on)Anthropic
76.0%
$15 / $75
78 Gemini 3.5 Flash-LiteGoogle
76.0%
$0.30 / $2.50
78 Gemini 3 Pro Preview (high reasoning)Google · best of 2 settings
76.0%
–
81 Inkling SmallThinking Machines
75.7%
$0.45 / $1.20
81 GLM-5Z.ai · best of 2 settings
75.7%
$0.95 / $2.55
83 Grok 4.3 (medium reasoning)xAI · best of 4 settings
75.0%
$1.25 / $2.50
84 DeepSeek-V4-Pro (0423, max reasoning)DeepSeek · best of 3 settings
74.7%
$1.42 / $2.83
84 o3OpenAI
74.7%
$2 / $8
84 Grok Build 0.1xAI
74.7%
$1 / $2
87 Qwen3-Max-ThinkingAlibaba
74.3%
$0.78 / $3.90
87 Claude Haiku 4.5 (reasoning on)Anthropic · best of 2 settings
74.3%
$1 / $5
87 DeepSeek-V4-Flash (0423, max reasoning)DeepSeek · best of 3 settings
74.3%
$0.14 / $0.28
87 Gemini 3.1 Flash-Lite PreviewGoogle
74.3%
$0.25 / $1.50
87 Gemini 3.5 Flash (medium reasoning)Google · best of 3 settings
74.3%
$1.50 / $9
87 MiMo-V2.6-FlashXiaomi
74.3%
$0.14 / $0.28
93 Grok 4.1 Fast (reasoning)xAI
74.0%
–
94 Grok 4 Fast (reasoning)xAI
73.7%
–
94 GLM-5.1Z.ai · best of 2 settings
73.7%
$1.38 / $4.40
96 DeepSeek-V3.2 (reasoning on)DeepSeek · best of 2 settings
73.3%
$0.30 / $0.96
96 MiniMax-M2.5MiniMax
73.3%
$0.30 / $1.20
98 Qwen3.7-PlusAlibaba
73.0%
$0.32 / $1.28
98 MiMo-V2.5Xiaomi
73.0%
$0.17 / $0.34
100 Claude Sonnet 4.5 (reasoning on)Anthropic · best of 2 settings
72.3%
$3 / $15
100 DeepSeek-V3.2-Exp (reasoning on)DeepSeek · best of 2 settings
72.3%
$0.27 / $0.41
100 GPT-5 mini (high reasoning)OpenAI · best of 3 settings
72.3%
$0.25 / $2
103 Qwen3-235B-A22B-Thinking-2507Alibaba
72.0%
$0.30 / $3
103 Qwen3.5-35B-A3BAlibaba · best of 2 settings
72.0%
$0.16 / $1.30
103 Kimi K2 ThinkingMoonshot AI
72.0%
$0.60 / $2.50
106 Qwen3.6-35B-A3BAlibaba · best of 2 settings
71.7%
$0.10 / $1
106 GPT-5-Codex (high reasoning)OpenAI
71.7%
–
106 GLM-5-TurboZ.ai
71.7%
$1.20 / $4
109 Gemini 2.5 Flash Preview (09-2025, reasoning on)Google · best of 2 settings
71.0%
–
109 GLM-4.7Z.ai · best of 2 settings
71.0%
$0.54 / $1.98
111 Solar Pro 4Upstage
70.7%
$0.09 / $0.36
112 Claude Sonnet 4 (reasoning on)Anthropic · best of 2 settings
70.3%
$3 / $15
112 GLM-5V-TurboZ.ai
70.3%
$1.20 / $4
114 Qwen3.5-9BAlibaba · best of 2 settings
70.0%
$0.10 / $0.15
114 DeepSeek-V3.2-SpecialeDeepSeek
70.0%
–
114 GPT-5.5 Instant (2026-06-26)OpenAI · best of 2 settings
70.0%
–
117 Gemma 4 31BGoogle · best of 2 settings
69.7%
$0.14 / $0.40
118 DeepSeek-V3.1-Terminus (reasoning on)DeepSeek · best of 2 settings
69.3%
$0.27 / $1
118 Mistral Medium 3.5Mistral AI
69.3%
$1.50 / $7.50
118 GPT-5.1-Codex (high reasoning)OpenAI
69.3%
$1.25 / $10
121 Gemini 2.5 ProGoogle
69.0%
$1.25 / $10
121 Nemotron 3.5 Super VLNVIDIA
69.0%
–
121 Grok 4.20xAI · best of 2 settings
69.0%
$1.25 / $2.50
124 GPT-4.1OpenAI
68.3%
$2 / $8
125 Grok 4xAI
68.0%
–
126 MiniMax-M2.1MiniMax
67.7%
$0.30 / $1.20
126 Grok 4.20 (0309, reasoning)xAI
67.7%
–
128 GPT-5.1-Codex-Mini (high reasoning)OpenAI
66.7%
$0.25 / $2
129 Gemma 4 26B A4BGoogle · best of 2 settings
65.7%
$0.10 / $0.30
129 Nemotron 3 Super 120B A12BNVIDIA
65.7%
$0.085 / $0.40
131 Gemini 2.5 Flash (reasoning on)Google · best of 2 settings
65.3%
$0.30 / $2.50
132 GPT-5 ChatOpenAI
65.0%
–
132 o1OpenAI
65.0%
$15 / $60
134 Gemini 2.5 Flash-Lite Preview (09-2025, reasoning on)Google · best of 2 settings
64.7%
–
134 Hy3 Preview (reasoning on)Tencent · best of 2 settings
64.7%
$0.18 / $0.60
136 MiniMax-M2MiniMax
64.3%
$0.30 / $1.20
137 Qwen3.5-Omni-PlusAlibaba
63.7%
–
137 Qwen3-Next-80B-A3B-ThinkingAlibaba
63.7%
$0.15 / $1.20
137 Qwen3-VL-235B-A22B-ThinkingAlibaba
63.7%
$0.40 / $4
137 Gemma 4 12BGoogle · best of 2 settings
63.7%
–
141 Qwen3.5-4B (reasoning on)Alibaba · best of 2 settings
63.0%
–
142 Qwen3-Max-Thinking-PreviewAlibaba
62.0%
–
143 Qwen3-30B-A3B-Thinking-2507Alibaba
61.3%
$0.20 / $2.40
144 o4-mini (high reasoning)OpenAI
61.0%
$1.10 / $4.40
145 Nova 2 Lite (high reasoning)Amazon · best of 4 settings
60.3%
$0.30 / $2.50
146 DeepSeek-R1DeepSeek
57.7%
$0.70 / $2.50
147 DeepSeek-V3.1 (reasoning on)DeepSeek · best of 2 settings
56.7%
$0.55 / $1.65
148 DeepSeek-R1-0528DeepSeek
55.7%
$0.50 / $2.18
148 Gemini 2.5 Flash-Lite (reasoning on)Google · best of 2 settings
55.7%
$0.10 / $0.40
150 Qwen3-VL-32B-ThinkingAlibaba
55.3%
–
151 GLM-4.6 (reasoning on)Z.ai · best of 2 settings
54.0%
$0.50 / $2
152 Magistral Medium 1.2Mistral AI
53.0%
–
152 Kimi K2 (0905)Moonshot AI
53.0%
$0.60 / $2.50
152 Kimi K2 (0711)Moonshot AI
53.0%
$0.57 / $2.30
152 Grok Code Fast 1xAI
53.0%
–
156 Qwen3-Next-80B-A3B-InstructAlibaba
52.7%
$0.10 / $1.10
156 GLM-4.5Z.ai
52.7%
$0.60 / $2.20
158 Qwen3.5-Omni-FlashAlibaba
52.0%
–
158 gpt-oss-120b (high reasoning)OpenAI · best of 2 settings
52.0%
$0.15 / $0.60
160 Qwen3-MaxAlibaba
50.0%
$0.78 / $3.90
160 Llama 4 MaverickMeta
50.0%
$0.27 / $0.85
160 Step 3.5 FlashStepFun
50.0%
$0.10 / $0.30
163 Mistral Small 4Mistral AI · best of 2 settings
49.7%
$0.15 / $0.60
164 GLM-4.6V (reasoning on)Z.ai · best of 2 settings
48.7%
$0.30 / $0.90
165 Qwen3-Coder-NextAlibaba
47.0%
$0.18 / $0.90
166 GLM-4.5-AirZ.ai
46.7%
$0.14 / $0.86
167 Qwen3-Coder-480B-A35BAlibaba
45.7%
$0.35 / $1.50
168 Granite 4.2 8BIBM
45.0%
$0.06 / $0.25
168 GPT-5 nano (high reasoning)OpenAI · best of 3 settings
45.0%
$0.05 / $0.40
170 GPT-4.1 miniOpenAI
44.0%
$0.40 / $1.60
171 Mercury 2Inception
43.7%
$0.25 / $0.75
172 Qwen3-Max-PreviewAlibaba
43.0%
–
172 o3-mini (high reasoning)OpenAI
43.0%
$1.10 / $4.40
174 Mistral Medium 3.1Mistral AI
42.7%
$0.40 / $2
175 GLM-4.7-FlashZ.ai · best of 2 settings
41.7%
$0.06 / $0.40
176 GPT-4o (2024-08-06)OpenAI
41.0%
$2.50 / $10
177 DeepSeek-V3-0324DeepSeek
40.7%
$0.25 / $1
178 Trinity Large ThinkingArcee AI
38.0%
$0.25 / $0.80
178 Nemotron 3 Nano 30B A3B (reasoning on)NVIDIA · best of 2 settings
38.0%
$0.05 / $0.20
180 Qwen3-4B-Thinking-2507Alibaba
37.3%
–
181 Mistral Large 3Mistral AI
36.0%
$0.50 / $1.50
182 gpt-oss-20b (high reasoning)OpenAI · best of 2 settings
34.7%
$0.03 / $0.15
183 Qwen3-235B-A22B-Instruct-2507Alibaba
33.9%
$0.15 / $0.75
184 Qwen3-VL-8B-ThinkingAlibaba
33.3%
$0.18 / $2.10
185 Qwen3-Coder-30B-A3B-InstructAlibaba
32.7%
$0.07 / $0.28
185 Qwen3-VL-235B-A22B-InstructAlibaba
32.7%
$0.30 / $1.50
187 Devstral 2Mistral AI
32.3%
$0.40 / $2
187 Solar Pro 3Upstage
32.3%
$0.15 / $0.60
189 Gemma 4 E4BGoogle · best of 2 settings
32.0%
–
189 Devstral MediumMistral AI
32.0%
–
191 Mistral Medium 3Mistral AI
31.3%
$0.40 / $2
191 Grok 4.1 Fast (non-reasoning, no reasoning)xAI
31.3%
–
193 DeepSeek-V3DeepSeek
29.3%
$0.26 / $1.03
194 Devstral Small 2Mistral AI
28.0%
–
194 Kimi Linear 48B A3B InstructMoonshot AI
28.0%
–
196 Llama 4 ScoutMeta
27.7%
$0.18 / $0.59
197 Qwen3-30B-A3B-Instruct-2507Alibaba
26.3%
$0.09 / $0.30
197 Ministral 3 14BMistral AI
26.3%
$0.20 / $0.20
199 Ministral 3 8BMistral AI
25.7%
$0.15 / $0.15
200 Grok 4.20 (0309, non-reasoning, no reasoning)xAI
25.3%
–
201 Grok 4 Fast (non-reasoning, no reasoning)xAI
24.0%
–
202 Mistral Small 3.1 24BMistral AI
22.3%
$0.35 / $0.56
203 Qwen3-VL-4B-ThinkingAlibaba
21.3%
–
203 Command ACohere
21.3%
$2.50 / $10
205 Qwen3.5-2B (reasoning on)Alibaba · best of 2 settings
21.0%
–
205 Nova Pro 1.0Amazon
21.0%
$0.80 / $3.20
207 Mistral Small 3.2 24BMistral AI
20.3%
$0.094 / $0.25
207 GPT-4.1 nanoOpenAI
20.3%
$0.10 / $0.40
209 DiffusionGemma 26B A4BGoogle
19.7%
–
210 Magistral Small 1.2Mistral AI
19.3%
–
211 Nova Lite 1.0Amazon
19.0%
$0.06 / $0.24
212 Devstral Small 1.1Mistral AI
18.7%
–
213 Llama 3.1 8B InstructMeta
18.0%
$0.05 / $0.08
214 Ministral 3 3BMistral AI
17.0%
$0.10 / $0.10
215 Qwen3-VL-8B-InstructAlibaba
16.7%
$0.12 / $0.46
216 Gemma 4 E2B (no reasoning)Google · best of 2 settings
16.3%
–
217 Llama 3.3 70B InstructMeta
15.7%
$0.59 / $0.79
218 Qwen3-VL-4B-InstructAlibaba
14.0%
–
219 Nova Micro 1.0Amazon
13.0%
$0.035 / $0.14
220 Qwen3-4B-Instruct-2507Alibaba
11.3%
–
221 Qwen3.5-0.8B (reasoning on)Alibaba · best of 2 settings
9.0%
–
222 Gemma 3 12BGoogle
8.3%
$0.05 / $0.15
223 Gemma 3 27BGoogle
7.3%
$0.12 / $0.20
224 Gemma 3 4BGoogle
6.7%
$0.05 / $0.10
224 Llama 3.2 1B InstructMeta
6.7%
$0.027 / $0.20
226 Llama 3.2 3B InstructMeta
4.3%
$0.05 / $0.33
227 Mistral Large 2 (2407)Mistral AI
2.0%
$2 / $6
228 Qwen3-14B (no reasoning)Alibaba · best of 2 settings
0.0%
$0.12 / $0.24
228 Qwen3-235B-A22B (no reasoning)Alibaba · best of 2 settings
0.0%
$0.46 / $1.82
228 Qwen3-30B-A3B (no reasoning)Alibaba · best of 2 settings
0.0%
$0.12 / $0.50
228 Qwen3-32B (no reasoning)Alibaba · best of 2 settings
0.0%
$0.14 / $0.40
228 Qwen3-8B (no reasoning)Alibaba · best of 2 settings
0.0%
$0.12 / $0.46
228 Qwen3-Omni-30B-A3B-InstructAlibaba
0.0%
–
228 Qwen3-Omni-30B-A3B-ThinkingAlibaba
0.0%
–
228 Gemma 3 270MGoogle
0.0%
–
228 Phi-4Microsoft
0.0%
$0.07 / $0.14
228 Mistral Small 3Mistral AI
0.0%
$0.05 / $0.08
228 Reka Flash 3Reka AI
0.0%
$0.10 / $0.20
228 GLM-4.5V (no reasoning)Z.ai · best of 2 settings
0.0%
$0.60 / $1.80

Swipe the table sideways for more columns.

Ranks follow the score as shown, so equal numbers share a rank. Each model is shown at its best setting; show every setting. Results as published by Artificial Analysis; we do not re-run them.

Progress over time

Long-context reasoning (AA-LCR), plotted at each model's release date, higher is better. Each model counts once, at its best setting. Scores are today's, not launch day's. Hover or tap a mark for its model.

Best score so far

Every model at its release date. The line moves each time a new release beats every earlier one, and each new best is named where there is room.

0%50%100% 2025Jul2026Jul Llama 3.1 8B InstructGPT-4o (2024-08-06)o1Gemini 2.5 Proo3Claude Opus 4.1GPT-5GPT-5.1GPT-5.2GPT-5.3-CodexGPT-5.5Kimi K3

Best score by lab

Each lab's best so far, stepping whenever one of its releases beats its own earlier models. The 6 labs with the best scores today; 19 others not drawn.

Moonshot AIXiaomiAnthropicOpenAIDeepSeekGoogle

0%20%40%60%80%100% 2025Jul2026Jul Moonshot AIXiaomiAnthropicOpenAIDeepSeekGoogle

What it measures

Share of questions answered correctly over long documents, as run by Artificial Analysis with its own harness and prompts.

What it does not measure

Not retrieval over your own document store.

Source

Artificial Analysis · Methodology

285 results from Artificial Analysis not ranked here · show why

We rank a result only when we can tie it to a specific model you can use. These are left out:

  • Not on sale through the API providers we track: 270
  • A different snapshot or variant from the model we list: 14
  • An unusual combination of settings: 1

Data sourced from Artificial Analysis. Licence: Artificial Analysis commercial data licence.